Artificial Intelligence Basics Explained with a Google Maps Example
Artificial intelligence (AI) is many things, among them a technology, an investment opportunity, and an intense subject of study by academics and experts from a wide variety of fields. If you want to learn about AI, you have to start somewhere, like using the laptop, desktop, or mobile device that you are using to read this article.
Artificial intelligence is very much about math and logic
Humans have intelligence that is based in part on biology and experience. Your mind takes in information, including lessons learned from your personal experiences and the experiences of others, and you make decisions and take actions based on the information and your judgment. It is a complex process, one so complex that many of the greatest minds in the history of the human race have not figured out exactly how the brain and the mind makes this happen. In recent decades, there has been great progress in emulating some parts of the process, particularly how to model the decision-making process using mathematics and logic and how to turn models into software-based tools that may be in the app on your phone or in a server used by some online service.
Artificial intelligence, at its core, is a decision-making tool for very specific tasks
As a human being, you use your intelligence many times a day to help you deal with mundane things like what clothes to wear that day or for more important things like whether to change careers. In many cases, there are many smaller decisions or actions that lie between you and your objective. To make those decisions, even the small ones, you have to do several things:
- Understand your needs and motivations,
- Decide what information you need to make a decision,
- Rely on your judgment and experience to both choose the data and make a decision,
- Choose a course of action (including taking no action at all) after making that decision,
- Understand the consequences for you and for others if the action has positive or negative consequences.
Artificial intelligence is commonly used to help make a decision
Artificial intelligence, whether in the laboratory or embedded in your mobile app, can’t do all of these things. Typically, they only do one small part of this process, making a prediction about some part of the decision process. In most practical applications of artificial intelligence used by businesses and individuals, artificial intelligence may only be applied to a specific task where three things are true:
- The prediction generated by the AI is better than the prediction a user makes without using that AI,
- The user can use that prediction to choose a course of action, and
- The user gets a positive benefit from that course of action.
Google Maps as an example of a practical AI in the real world
One of the most popular kinds of tools are mapping applications like Google Maps. Google Maps takes some user input and then provides the user with predictions about travel times for one or more potential routes, and highlights what the Google Maps algorithm predicts is the option that gives the user the greatest value. In the example below, a user accessed Google maps to show the routes from a shopping mall in Burlington, MA on the outskirts of the Boston metropolitan area to Harvard Square in Cambridge, MA. The map shows three options for that trip, including the estimated time to complete each option, and the trip. The route preferred by Google Maps, the one with the lowest estimated travel time, is in blue.
A user has to choose an action after Google Maps makes a prediction
The Google Maps prediction of possible travel times helps the user only with the tasks of choosing a route. The user has to then use his or her judgment to decide which option (including the option to take no action) is best for the user. Typically that choice depends on information that Google Maps does not have and can’t predict. For example:
- The route that takes the longest time passes by the user’s favorite lunch spot,
- The next longest route has very relaxing scenery, or
- The user decides at the last minute to blow off the trip and hang out in the mall.
Artificial Intelligence operates in the background
When the user picks a route using Google Maps, there are likely dozens of AI-related activities taking place in the background. The user doesn’t understand and does not have a need to understand exactly how aprobably does not care about what is going on inside the software in the user’s mobile device or in the Google Maps servers that analyze all the data. All a user cares about is whether the technology works.
Artificial intelligence is a lot more than Google Maps
Artificial intelligence isn’t new. It’s been studied by researchers in government, industry, and academia for decades. What has changed in the last decade or so is the widespread use of portable devices like mobile phones, and applications designed for those devices that use artificial intelligence and other data-related technologies to give users decision making tools of all kinds. If you are doing something online that was either difficult or time consuming ten or twenty years ago, like booking a flight or a hotel room, and is less time consuming and less likely to involve a human being, then some kind of AI is probably working the background.